이번 학기 콜로퀴움은 학교의 권고를 따라, 사회적 거리두기 단계와 연계하여 온라인 혹은 오프라인(온라인 방송 병행)으로 운영될 예정입니다.

첫번째 강연에는 성균관대학교 소프트웨어학과 박진영 교수님을 모시고 콜로퀴움을 개최하오니, 학과 구성원 여러분의 많은 참여를 부탁 드립니다.

코로나19 확산방지를 위해 다과는 제공하지 않습니다.

                                      아            래

1. Title: Scaling laws in Natural Language Processing

2. Speaker: 박  진  영  교수님 (성균관대학교 소프트웨어학과)

3. Date & Time: Mar. 3. (Wed.) 2021.  4:30 PM

4. Place: Natural Science 1,  Room No. 31214

5. Abstract: Many artificial intelligence (AI), machine learning (ML), and natural language processing (NLP) models have seen successes in recent years. Based on these models, there are many good NLP applications such as machine translation, question answering and chatbot. But, we still have questions about the models. One of the questions is the relationship between model performance and several factors such as model size.
Scaling law (power-law) is a functional relationship between two quantities and has been identified in physics. It is also observed in NLP too. In this talk, I will introduce a language model that is an important model for NLP and the power-law distribution between the factors and performance of the model. I will also briefly introduce the convergence of ML and physics.
 

Bio:
JinYeong Bak is an assistant professor in the colleague of computing at Sungkyunkwan
University. His research interests analyzing human behaviors from their conversations and building open-domain conversation models from the insights of the analysis. He worked at Microsoft Research Asia as a research intern and United Nations Global Pulse Lab Jakarta as a junior data scientist. He received Ph.D. and M.S. from the KAIST and a B.S. from Sungkyunkwan University. His research has been published in ACL, EMNLP, CHI, and ICWSM. His personal homepage: https://nosyu.github.io and lab homepage: https://hli.skku.edu